jpk-korekta

jpk-korekta is a skill for Claude Code from dithiothreitol/jdg-ksiegowy. It costs 160 tokens per session (1,163 once invoked), scanned A, original, MIT.

A tool for correcting a Polish JPK_V7M filing for a chosen month. JPK_V7M is an electronic Polish VAT report combining the tax declaration and sales and purchase records.

In plain words
What is it for?
It generates a correction file, adds the correction reason when needed, validates it against the official XSD schema, and sends it for submission.
Why use it?
It helps replace an already submitted report when the VAT filing is wrong, while checking the required XML format and submitting the correction to the Polish Ministry of Finance gateway.

Skill for Claude Code

Written for Claude Code: installed under .claude/.

Needs its repository: it runs a file that does not travel with it, so clone the repository first. The line is python3 skills/jpk/scripts/generate_jpk.py --month 3 --year 2026 --correction \.

Good fit It generates a correction file, adds the correction reason when needed, validates it against the official XSD schema, and sends it for submission.

Compare 6 skills from other repositories ↓
Install

Getting it into your agent

It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.

Clone the repo
git clone --depth 1 https://github.com/dithiothreitol/jdg-ksiegowy
agentmods
npx agentmods add skills/dithiothreitol/jdg-ksiegowy/jpk-korekta

Made for: Claude Code.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for jpk-korekta

README.md
[![agentmods](https://agentmods.dev/badge/skills/dithiothreitol/jdg-ksiegowy/jpk-korekta/github.svg)](https://agentmods.dev/skills/dithiothreitol/jdg-ksiegowy/jpk-korekta)
Your own site
<a href="https://agentmods.dev/skills/dithiothreitol/jdg-ksiegowy/jpk-korekta"><img src="https://agentmods.dev/badge/skills/dithiothreitol/jdg-ksiegowy/jpk-korekta/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for jpk-korekta

Your own site · 80×15
<a href="https://agentmods.dev/skills/dithiothreitol/jdg-ksiegowy/jpk-korekta"><img src="https://agentmods.dev/badge/skills/dithiothreitol/jdg-ksiegowy/jpk-korekta.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 160 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,163 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00160 $0.01163
Opus 5.5 $0.00064 $0.00465
Sonnet 5.5 $0.00032 $0.00233
Haiku 4.5 $0.00016 $0.00116

Measured 5d ago against content hash eeb9bf573f8b, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

jpk-korekta scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 5d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

.claude/skills/jpk-korekta/SKILL.md · 56 lines

How it starts

The opening of the file, as written. The whole thing — 56 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Korekta JPK_V7M

Od 10.2020 deklaracja VAT-7 nie istnieje osobno — jest częścią JPK_V7M. Korekta to pełny plik JPK_V7M za ten sam okres z CelZlozenia = 2 (formularza ORD-ZU nie ma; uzasadnienie to opcjonalne pole P_ORDZU). W urzędzie obowiązuje ostatni przyjęty plik (najnowsze UPO) — dlatego zerowa deklaracja wysłana przez pełnomocnika po Twojej nadpisuje Twoje rozliczenie, a Twoja korekta nadpisuje ją z powrotem.

Kiedy korekta, a kiedy nie

  • Za okres istnieje jakikolwiek przyjęty plik (Twój albo pełnomocnika) → korekta (--correction).
  • Za okres nikt nic nie złożył → zwykłe złożenie (generate_jpk.py bez flagi); bramka może odrzucić korektę bez pierwotnego.
  • Nie wiesz, co jest w urzędzie → user sprawdza w e-Urzędzie Skarbowym (Dokumenty → JPK_VAT). Nie zgaduj — brak UPO w data/upo/ nie znaczy, że za okres nic nie ma.

Przebieg (dla każdego okresu)

# 1. Korekta z uzasadnieniem — oryginał data/jpk/JPK_V7M_2026_03.xml zostaje, powstaje _K1.xml
python3 skills/jpk/scripts/generate_jpk.py --month 3 --year 2026 --correction \
  --reason "Korekta przywraca prawidłowe rozliczenie za okres. Pełnomocnik (biuro rachunkowe) omyłkowo złożył za ten okres deklarację zerową."

# 2. Walidacja XSD (bramka MF nie jest walidowana lokalnie — błąd wyszedłby dopiero jako 4xx)
python3 skills/jpk/scripts/validate_jpk.py --xml-path data/jpk/JPK_V7M_2026_03_K1.xml

# 3. Wysyłka jak zwykłego JPK (skill jpk-submit) — najpierw --dry-run
python3 skills/jpk-submit/scripts/submit.py --xml-path data/jpk/JPK_V7M_2026_03_K1.xml --dry-run
python3 skills/jpk-submit/scripts/submit.py --xml-path data/jpk/JPK_V7M_2026_03_K1.xml

Kolejna korekta tego samego okresu → _K2.xml (numer nadawany automatycznie); w XML zawsze CelZlozenia=2.

Przed wysyłką pokaż userowi

Tabelę per okres: P_38 (należny), P_48 (naliczony), P_51 (do zapłaty) z korekty vs. z oryginalnie złożonego pliku (data/jpk/JPK_V7M_YYYY_MM.xml). Różnica = rejestr zmienił się od złożenia (np. dopisane koszty) — user musi świadomie zaakceptować, że korekta wysyła nowe kwoty. Pokaż też treść P_ORDZU.

Read the full file on GitHub · 56 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 5d ago First seen · 56 lines · 160 tokens per session scan A eeb9bf573f8b

Subscribe to this mod's changes

jpk-korekta is a skill published in the GitHub repository dithiothreitol/jdg-ksiegowy (9 stars, last pushed 5d ago), licensed MIT. It adds 160 tokens to every session and 1,163 once invoked, about $0.0006 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-10-21.

Related

Other skills, from other repositories

jangbu-jongso

A Korean tax-document checklist generator for individuals, freelancers, sole-company owners, mixed earners, and certain business owners.

kimlawtech/korean-jangbu-for · 139 tokens

brazil-vat

> Use esta skill sempre que perguntarem sobre tributos indiretos brasileiros, IVA, tributos sobre o consumo, PIS, Cofins, ICMS, ISS, IPI, CBS, IBS, Imposto Seletivo, NF-e (Nota Fiscal Eletronica), NFS-e, Simples Nacional, registro de CNPJ, reforma tributaria (EC 132/2023, LC 214/2025, LC 227/2026) ou qualquer questao…

openaccountants/openaccountants · 213 tokens

az-payroll

Tier 2 Arizona content skill for employer payroll compliance covering tax year 2025. Includes the 2.5% flat PIT (phased down from 4.5% over 2022-2023), A-4 state W-4 expressed as percentage of federal withholding (0.5-3.5% options), A1-QRT quarterly withholding, A1-R annual reconciliation, AZ UI wage base $8,000 with…

openaccountants/openaccountants · 132 tokens

bb-income-tax

Use this skill whenever asked about Barbados personal income tax, NIS contributions, or payroll obligations. Trigger on phrases like "how much tax do I pay in Barbados", "BRA", "Barbados Revenue Authority", "TAMIS", "PAYE Barbados", "NIS contributions", "National Insurance Barbados", "self-employed tax Barbados"…

openaccountants/openaccountants · 193 tokens

at-income-tax

Austrian income tax (Einkommensteuer) for resident individuals, tax year 2026: employees, pensioners, sole traders, freelancers and private landlords. Covers the 2026 bands (0/20/30/40/48/50/55%) as indexed by two thirds of 2.6% inflation, the credits (Verkehrsabsetzbetrag, sole-earner and single-parent, pensioner…

openaccountants/openaccountants · 278 tokens

bulgaria-vat-return

Use this skill whenever asked to prepare, review, or classify transactions for a Bulgarian VAT return (Spravka-deklaratsiya po DDS) for any client. Trigger on phrases like "prepare VAT return", "do the DDS", "fill in DDS return", "create the return", "Bulgarian VAT", or any request involving Bulgaria VAT filing. Also…

openaccountants/openaccountants · 158 tokens